Case Study: Singapore’s Sugar Beverage Health Strategy

By Hao Yi Tan

Sugary Drinks and The Rise of Chronic Disease

In many parts of the developed world, populations struggle with a rise in chronic diseases such as hypertension, diabetes, high cholesterol, and obesity. These medical conditions have high morbidity, and are often pre-cursors to more issues such as ischaemic heart disease, stroke and kidney failure. These chronic diseases cause an increased pressure on the healthcare system. As such, the state has a strong incentive to enact policies to reduce the prevalence of such diseases to avoid an overwhelming burden on the healthcare system.

There are many risk factors for the aforementioned chronic diseases, including but not limited to genetic predisposition, family upbringing, and sedentary lifestyle. However, scientific evidence points towards unhealthy dietary intake being a key driver of these diseases. For example, high consumption of sugar-rich beverages increases the risk of developing type 2 diabetes mellitus. In a perfect market, consumers are able to make well-informed decisions based on the risks of what they consume. However, advertising efforts from companies attempting to market their products as “healthy” can often misrepresent facts and mislead consumers. Furthermore, individuals may lack the information to know which foods are healthy or unhealthy for them. This information mismatch has a direct impact on their health and the health of their dependents (e.g. children, elderly parents). As such, some governments have tried to influence dietary choices by using food or beverage labelling policies.

Singapore’s Food Labeling Strategy

Figure 1: Nutri-Grade Grading Logo (Source: Singapore Health Promotion Board)

One policy that aims to combat the problem of dietary choice is Singapore’s Nutri-Grade Mark policy, which specifically focuses on sugar-sweetened beverages. The Nutri-Grade policy was officially implemented on 30 December 2022, affecting all stores selling beverages including supermarkets, convenience stores, food and beverage (F&B) establishments and automated beverage dispensers (1).

According to Singapore’s National Nutrition Survey 2018/2019 (2), Singaporeans, on average, consume a daily amount of twelve teaspoons (equivalent to 58g) of sugar. This is significantly higher than the daily recommended amount of 25-36g (3). Notably, Singapore has the highest prevalence of diabetes amongst developed nations (4). The majority of their daily sugar intake is derived from beverages, with pre-packaged beverages contributing 64% and freshly prepared beverages contributing 36%. The government therefore saw a pressing need to decrease sugar consumption among Singaporeans.

Singapore’s Health Promotion Board (HPB) stated that the policy’s main aim was to help consumers identify beverages with higher sugar and saturated fat content. In doing so, they aim to “reduce the influence of advertising on consumer preferences, thus encouraging more informed, healthier choices and spurring industry reformulation”. The Nutri-Grade label ‘grades’ the beverage on a score of A to D, with A being the healthiest (see figure 1 below). The grading takes into account sugar content and saturated fat content (measured in grams per 100ml).

Figure 2: Nutri-Grade Grading System (Source: Singapore Health Promotion Board)

Nutri-Grade labelling was made mandatory for all beverages which falls into the C and D grading, and optional for those that fall into the A and B grading. A and B gradings also have the option to place a “healthier choice” label on their products (a food labelling policy that predates Nutri-Grade). Additionally, beverages with a D grading will have advertising prohibitions, disallowing sellers from actively promoting them. All beverages are required to display nutrition information regardless of Nutri-Grade.

This policy also aims to decrease disparities through simplified scoring of beverages. For Singaporeans not familiar with nutrition and food science, ascertaining whether the drink they are buying is healthy or not from the ingredient list is difficult. In Singapore, individuals with lower levels of education are generally those with lower socio-economic status (5) and the elderly (6). Both these groups are less likely able to discern healthier beverages or know about the consequences of constant consumption of such beverages. These groups also have disproportionately higher rates of chronic diseases (7, 8), making them the natural target populations for this policy.

Intended Policy Mechanism

The intended mechanism of this policy is to provide an aggregated scoring system to inform consumers of how healthy beverages are at the point of sale. The hope is that Singaporeans, empowered with dietary knowledge will make better purchase decisions in the beverages that they buy, diminishing a major source of sugar and saturated fat in their overall diet. This, along with other health promotion campaigns aim to prevent overconsumption of food substances that increase the risk of chronic diseases. In turn, this will hopefully decrease their prevalence and burden on the healthcare system, while enabling Singaporeans to remain healthy into their old age.

As stated by HPB, a secondary mechanism is to increase demand for healthier beverages through this labelling policy, resulting in a corresponding decrease in demand for unhealthy ones. In this situation, market forces naturally incentivize firms to innovate and sell more healthy beverages, creating a virtuous cycle that ultimately benefits Singaporeans.

Has Singapore’s Strategy Worked?

As of Jan 2024, there has only been one experimental study (9) published that examined this specific policy. In this study published in February 2023, researchers conducted an experimental study with a two-arm crossover design to determine the effectives of Nutri-Grade labelling. They recruited these participants on Instagram and Facebook and email blasts from the research team. Participants were only eligible if they were above 21 years of age and were the primary grocery shoppers for their family. 138 eligible participants were randomized and placed in 2 groups, one with the Nutri-Grade labels, and one without (control). They were asked to shop virtually as they normally would, simulating an actual shopping trip. There was also 50% chance that they will actually have to pay for their shopping at the time of the experiment. This was an attempt by researchers to minimize the impact of ‘hypothetical shopping’ or observer bias because they had a significant chance of needing to pay for the items. After this, each group was given a 7-day break then were switched over to the opposite group.

The results were able to show that overall, the Nutri-Grade label was effective at increasing the rate of healthier purchases (as defined by Nutri-Grade A and B products) by a statistically significant amount compared to the controls. However, the effect was only limited to sugar content, as there was no difference in the amount of saturated fat and total calories of beverages purchased. Buyers who had no household members with diet-related health conditions were ironically more likely to purchase Nutri-Grade A and B beverages compared to those with household members with such conditions. The study also found that shoppers who had high Body Mass Index (BMI) were more likely to buy high-sugar beverages than those with normal BMI. In terms socioeconomic status, high-income shoppers (earning SGD$10,000 and above) had a greater mean reduction in purchased sugar per serving in response to Nutri-Grade labelling compared to low-income shoppers.

These results seem to indicate modest success of the policy’s ability to shape consumer behaviour. However, the study design has some notable issues. Firstly, the crossover design was chosen to reduce the number of individuals needed to be recruited by researchers. However, crossover designs run the risk of suffering from contamination, possibly affecting internal validity of the study.

Furthermore, participants were recruited through social media and email, which causes a selection bias by only choosing individuals who are technologically savvy. Elderly individuals who are not digitally connected will not have been recruited for the study, even if they met all criteria. Individuals below the age of 21 and those who were not main shoppers for their household were also excluded. However, these are a non-trivial group of shoppers. Children for example, have been clearly shown to affect purchasing habits of parents (10), and including them may produce differing results. Additionally, a significant number of individuals might solely shop for beverages, without being the main shoppers for the household. These indicate a possible lack of external validity of this study.

Similar Policies in Other Countries

Similar policies have been studied in other developed nations like Australia, New Zealand, Canada and France (11-13).

The closest in terms of intervention intent and studied outcomes is published by Acton and Hammond in 2018, investigating the effects of price and nutrition labelling on sugary drink purchases. This experimental marketplace study in Canada studied 675 participants, where they were randomized to one of four labelling conditions (no label, star rating, high sugar symbol, and health warning). Participants were tasked with completing five “within-subject” purchases from a selection of 20 commercially available beverages. Beverage prices in each task corresponded to “tax” conditions of 0%, 10%, 20%, 30% and a variable tax corresponding to free sugar level.

This study found that increasing price was associated with reduced likelihood of purchasing sugary beverages, resulting in selection of drinks with less calories and less sugar content. Overall, labelling of beverages did reduce likelihood of individuals selecting a sugary drinks and encouraged participants to select drinks with less free sugar, but this was not statistically significant. 

Empirical Evidence for Food Labeling Policies

A 2015 meta-analysis (14) of 10 different food labelling studies concluded that food labelling may play a significant role in facilitating the selection of healthier foods. Specifically, they pointed out that the traffic light schemes were the most effective method in steering consumers’ dietary choices. One possible hypothesis being that nutrient information might be too confusing for customers, whereas a simple colour-coded system was more understandable and therefore more useful.

Note: Traffic light labels denote the levels of four key ingredients (i.e., fat, sugar, saturates, and salt) commonly contained in processed food, with red indicating a high level, amber a medium level, and green a low level of the respective nutrient (15).

However, they note that the findings show a less clear picture in terms of whether food labelling policies actually lower unhealthy consumption, such as ‘total calories consumed’, since none of those results were statistically significant. The authors note that since most of the studies report large confidence intervals, it suggests that individuals’ response to the food labels vary widely, ranging from indifference to a significant change in purchasing.

This meta-analysis notes that the majority of the studies included were carried out in controlled settings (like the aforementioned Singapore study) or as online trials. The results shown might therefore not be completely representative of the real-world results in supermarkets and stores. The lack of external validity hence indicates a need for more research to be done in less sterile real-world scenarios to truly determine effectiveness of such interventions.

Another 2019 meta-analysis (16) of 60 papers on the effects of food labelling on consumer diet behaviors and industry practice found that it reduced consumer consumption of total energy and total fat, while increasing consumption of vegetables. It also found that food labelling altered industry formulations for sodium and trans-fat, but did not significantly affect product formulations for total energy, saturated fat, dietary fiber, or other dietary components (whether healthy or unhealthy). Notably, studies included in this analysis were mainly natural experiments and not randomized trials, increasing the likelihood that these results are more generalizable in a real-world scenario.

Summary

In conclusion, the implementation of Singapore's Nutri-Grade Mark policy represents a significant step towards addressing the rising prevalence of diabetes linked to unhealthy beverage choices. Moreover, the policy implicitly helps populations with lower health literacy and socioeconomic status, given their relative vulnerability to misinformation and higher rates of chronic diseases.

While the experimental study conducted in Singapore suggests a positive impact on purchasing behavior, it is essential to acknowledge the study's limitations, including a potential lack of external validity. Similar policies in other developed nations have shown mixed results too, emphasizing the need for more research in real-world scenarios. Meta-analyses indicate that food labelling, particularly schemes like the traffic light system, can play a role in guiding healthier choices. However, the effectiveness of these interventions in significantly reducing unhealthy consumption is still not certain.

In moving forward, it is crucial to consider the broader implications of food labelling policies and their potential to create a virtuous cycle in the market, encouraging industry reformulation and fostering a culture of healthier choices. Continuous research and studying policies in the real-world setting will be key to refining interventions that truly impact consumer behavior and contribute to a healthier society.

---

References

1. Health Promotion Board (HPB) Singapore. Measures for Nutri-Grade Beverages 2023 [Available from: https://hpb.gov.sg/healthy-living/food-beverage/nutri-grade.

2. Health Promotion Board (HPB) Singapore. National Nutrition Survey 2018 Shows Gradual Improvements in Singaporeans’ Dietary Habits: Health Promotion Board (HPB) Singapore,; 2019 [Available from: https://www.hpb.gov.sg/newsroom/article/national-nutrition-survey-2018-shows-gradual-improvements-in-singaporeans-dietary-habits.

3. American Health Association (AHA). How much sugar is too much? : American Health Association (AHA),; 2024 [Available from: https://www.heart.org/en/healthy-living/healthy-eating/eat-smart/sugar/how-much-sugar-is-too-much#:~:text=To%20keep%20all%20of%20this,or%20100%20calories)%20per%20day.

4. Ministry of Health Singapore. PUBLIC CONSULTATION ON MEASURES TO REDUCE SUGAR INTAKE FROM PRE-PACKAGED SUGAR-SWEETENED BEVERAGES: Ministry of Health Singapore,; 2018 [

5. Mukhopadhaya P. Education policies as means to tackle income disparity: the Singapore case. International Journal of Social Economics. 2002;29(12):946-55.

6. Subramaniam M, Abdin E, Vaingankar JA, Sambasivam R, Seow E, Picco L, et al. Successful ageing in Singapore: prevalence and correlates from a national survey of older adults. Singapore medical journal. 2019;60(1):22.

7. Tan ST, Quek RYC, Haldane V, Koh JJK, Han EKL, Ong SE, et al. The social determinants of chronic disease management: perspectives of elderly patients with hypertension from low socio-economic background in Singapore. International journal for equity in health. 2019;18(1):1-14.

8. Subramaniam M. Disability in Singapore’s elderly population. Ann Acad Med Singapore. 2016;45:284-96.

9. Shin S, Puri J, Finkelstein E. A randomized trial to evaluate the impact of Singapore’s forthcoming Nutri-grade front-of-pack beverage label on food and beverage purchases. International Journal of Behavioral Nutrition and Physical Activity. 2023;20(1):18.

10. Baldassarre F, Campo R, Falcone A. Food for kids: How children influence their parents purchasing decisions. Journal of Food Products Marketing. 2016;22(5):596-609.

11. Acton RB, Hammond D. The impact of price and nutrition labelling on sugary drink purchases: Results from an experimental marketplace study. Appetite. 2018;121:129-37.

12. Ni Mhurchu C, Volkova E, Jiang Y, Eyles H, Michie J, Neal B, et al. Effects of interpretive nutrition labels on consumer food purchases: the Starlight randomized controlled trial. The American journal of clinical nutrition. 2017;105(3):695-704.

13. Chantal J, Sandrine P, Camille B, Rebeca G, Mathilde T, Serge H, et al. Perception of different formats of front-of-pack nutrition labels according to sociodemographic, lifestyle and dietary factors in a French population: cross-sectional study among the NutriNet-Santé cohort participants. BMJ Open. 2017;7(6):e016108.

14. Cecchini M, Warin L. Impact of food labelling systems on food choices and eating behaviours: a systematic review and meta‐analysis of randomized studies. Obesity reviews. 2016;17(3):201-10.

15. Kunz S, Haasova S, Rieß J, Florack A. Beyond healthiness: the impact of traffic light labels on taste expectations and purchase intentions. Foods. 2020;9(2):134.

16. Shangguan S, Afshin A, Shulkin M, Ma W, Marsden D, Smith J, et al. A meta-analysis of food labeling effects on consumer diet behaviors and industry practices. American journal of preventive medicine. 2019;56(2):300-14.